Research on energy-saving optimization of EMU trains based on golden ratio genetic algorithm

In order to study EMU (electric multiple units) trains operation control with attention to minimizing the energy consumption, a multi-particle method to deal with additional resistances was proposed aiming at the problem that the force analysis of single-particle model for the train was inaccurate, and two optimizations were carried out based on the multi-particle model. Then, a method with golden ratio genetic algorithm was proposed to solve the problem that genetic algorithm was easy to fall into local optimum, by which a set of target speed sets satisfying constraints were sought for the train in the first optimization, thus the train energy-saving operation speed curve was determined. Considering the influence of electrical phases for the train operation, the second optimization was carried out. The operation interval was divided into fixed segments and optimizable segments of manipulation, and a set of satisfactory operation switching points were searched by golden ratio genetic algorithm. The final operation curve of the train was obtained in tandem with the first optimization. Taking CRH3 of Lankao South-Kaifeng North line as a simulation case, the energy consumption of the train operation is reduced by 10.83%, which shows that the proposed method is feasible. © 2020, Central South University Press. All rights reserved.